FridaSAFE

English | 中文

English

A Model Context Protocol (MCP) server that enables AI models to perform Android dynamic analysis using Frida.

Project Structure

frida-mcp/
├── frida_mcp.py          # Core MCP server implementation
├── pyproject.toml        # Project dependencies and configuration  
├── requirements.txt      # Alternative dependency file
├── config.json           # Optional Frida server configuration
├── README.md            # Documentation
└── .gitignore           # Git ignore rules

Core Files

• `frida_mcp.py`: Main MCP server with Frida integration
• `pyproject.toml`: Modern Python project configuration (recommended)
• `requirements.txt`: Traditional dependency file
• `config.json`: Optional configuration for frida-server settings

Installation

# Clone repository
git clone http://git.dev.sh.ctripcorp.com/octopus/frida-mcp.git
cd frida-mcp

# Install dependencies (choose one method)
# Method 1: Using pip + requirements.txt
pip install -r requirements.txt

# Method 2: Using pip + pyproject.toml (recommended)
pip install -e .

# Setup frida-server on Android device
adb push frida-server /data/local/tmp/
adb shell "chmod 755 /data/local/tmp/frida-server"
adb shell "su -c /data/local/tmp/frida-server &"

MCP Configuration

Add to your MCP client configuration (e.g., Claude Desktop config file):

Windows (%APPDATA%\Claude\claude_desktop_config.json):

{
"mcpServers": {
"frida": {
"command": "python",
"args": ["C:\\Users\\YourName\\frida-mcp\\frida_mcp.py"],
"transport": "stdio"
}
}
}

macOS/Linux (~/.config/claude/claude_desktop_config.json):

{
"mcpServers": {
"frida": {
"command": "python",
"args": ["/home/username/frida-mcp/frida_mcp.py"],
"transport": "stdio"
}
}
}

Configuration (Optional)

The config.json file contains optional Frida server config